Home

The CHIL Knowledge Base Server

As the underlying Semantic Web standards such as RDF(S), DAML+OIL, and their common Description Logics (DL) based successor OWL DL have matured, tools for ontology engineering have emerged both in commercial as well as in academic fields. Knowledge acquisition systems such as Protégé make it particularly easy to construct domain ontologies and to enter data. Ontology management systems such as the Jena Semantic Web Framework can be used for loading OWL DL ontologies from files and via the Internet and for creating, modifying, querying, and storing ontologies. Inference engines such as Pellet provide support for query answering. There is a growing set of tools, projects, and applications for the SHOIN(D) Description Logic based Web Ontology Language (OWL DL). However, processing ontological information using existing off-the-shelf ontology management systems is still laborious and error-prone. This is inter alia caused by two main problems.

• Firstly, there are no formal specifications that fully define the semantics of ontology management APIs. This is particularly problematic since typical interface methods of OWL APIs (e.g., listSubclasses, addIndividual) are closely related to the semantics of the formal foundations (i.e. Description Logics) of the Web Ontology Language.

• Secondly, existing off-the-shelf ontology management systems provide only limited connectivity with respect to native support for programming languages and remoting protocols. Hence, it is particularly difficult to use ontology management systems remotely or along with a variety of different programming languages (i.e. in heterogeneous distributed computing environments). More severely, it may be unfeasible to replace an ontology management system by alternative products without rewriting significant parts of client code.

The CHIL Knowledge Base Server (CHIL KBS) is an ontological knowledge base server with a pluggable architectural model. The CHIL KBS can expose the ontology management functionality of arbitrary off-the-shelf OWL DL systems via a well defined API. A combination of Description Logics terminology and Floyd-Hoare logic was used to formally specify the result sets and side effects of each method.

The architectural model of the CHIL Knowledge Base Server facilitates adding and replacing remoting protocol hosts and replacing the adapter code for arbitrary off-the-shelf OWL DL management systems. Implementations of the knowledge base server API can automatically be tested for adherence to the formal specification using a regression test framework. Client libraries are available for a number of programming languages.

See the XML Schemas page for information about how to configure a CHIL KBS instance.

The CHIL Knowledge Base Server was actively used in the CHIL research project.